Gene Identification Using the Pufferfish, Fugu rubripes, by Sequence Scanning

The major goal of the Human Genome Project must be to identify, sequence, characterize and assign specific function to all the genes spread through our 3000 Mb of haploid DNA. Owing to the uniformity and simplicity of the DNA code, it is not an easy task to identify genes even after the region in which they lie has been fully sequenced. The average length of a human coding sequence in the DNA databases is approx 1.2 kb, and sensible estimates of the total number of genes in the human genome lie between 50 and 100,000. A gene number of 70,000 would give a total coding sequence of 85 Mb, <3% of our genome. Herein lies one of the major problems. A 3% return on investment, even when genes are identifiable, is rather poor, especially when sequencing is an expensive business. As if that is not enough, a large percentage of highly reiterated dispersed repeats serves to exacerbate the problem.